Democratic Values and Support for Militancy:

Evidence from a National Survey of Pakistan

C. Christine Fair
Georgetown University
Neil Malhotra
University of Pennsylvania
Jacob N. Shapiro
Princeton University
February 8, 2011

Abstract
A long tradition of research into political culture argues that greater support for core liberal
democratic values leads to a rejection of destructive political activities and reduced support for
violent politics. Policymakers have long drawn on this line of enquiry, arguing that exporting
democracy can reduce violent political activity such as terrorism. Unfortunately, there have been few
direct tests of the hypothesis that mass-level support for democratic values correlates with the
rejection of violence. We conduct such a test in Pakistan, a country that is both a significant source
of militancy and one whose citizens suffer massively from terrorism. We designed, conducted, and
analyzed an original 6,000-person provincially-representative survey and find that strong supporters
of democratic values are actually more supportive of militant groups focused outside of Pakistan than
Pakistanis less favorable towards liberal democracy. Consistent with the principle of azadi, this result
is driven by those who believe that Muslim rights and sovereignty are being violated in Kashmir.
Our results both challenge the conventional wisdom that has formed the basis of public policy as
well as contribute to theoretical debates on the influence of civic culture on political stability and
violence.

A long tradition of research into political culture has suggested that greater support for core
liberal democratic values leads to a rejection of destructive political activities and produces a wide
range of benefits from resistance to autocratic coups to greater economic growth (e.g. Almond and
Verba 1963; Powell 1982; Huntington 1984; Dalton 1994; Gibson 1997; Traube 2008; Kirwin and
Cho 2009; Persson and Tabbelini 2009). Scholars have posited numerous mechanisms that may
underlie this posited relationship. For example, mass support for democratic values could induce
citizens to peacefully work within their political system and engage in collective action to reject
uprisings and preserve civil society. In the case of Pakistan, individuals committed to core liberal
values may reject the vision offered by Islamist militants who seek to enforce their (often stylized)
version of Sharia law through violence.
Ostensibly drawing from this line of argumentation, a major tenet of U.S. foreign policy
under the Bush administrationand one that still holds sway among many in the Obama
administrationis that exporting democracy to regions of the world where it is absent may
reduce support for violent political activity such as terrorism (see e.g. National Security Council
2006, Hamid and Brooke 2010). What has been missing from this discussion is individual-level data
assessing whether democratic values relate to support for violent political organizations.
A thoughtful reflection upon the claims made by many militant groups over the last fifty
years and on the nature of competition between governments and militant groups in many
regionssuggests that theories about the palliative role of democratic values need to take careful
consideration of political context. Since at least the American Revolution there has been no shortage
of political movements that have rallied their followers to kill and risk death in the name of freedom,
democratic representation, and other liberal democratic values. In South Asia and the Middle East
today, many militant groups claim to be defending freedom, fighting for self-representation, and
mobilizing against what they perceive to be corrupt governments. Moreover, the populations from
1

which these varied militant groups draw support appear to believe these claims. Accordingly,
support for liberal democratic values may actually be positively correlated with support for militancy,
and may be particularly dependent on peoples reactions to the political environment. Unfortunately,
there exists scant evidence to test this conjecture.
In this paper, we empirically test the hypothesis that support for core democratic values is
associated with a rejection of violent politics. We examine Pakistan, a country that is both a
significant source of militancy and one whose citizens suffer massively from terrorism. We designed,
conducted, and analyzed an original 6,000-person survey that is representative of adults in each of
Pakistans four main provinces: Punjab, Sindh, Khyber-Pakhtunkwa (KPK), and Balochistan. Our
survey is the first to measure affect towards a range of specific militant organizations in one country,
the first to measure respondents beliefs about the importance of core democratic values, and the
first to be representative of rural and urban regions for each province of Pakistan.
Most importantly, we apply a novel measurement strategy to mitigate item non-response and
social desirability bias, which have plagued previous surveys due to the sensitive nature of militancy
in the region. 1 Our Endorsement Experiment (detailed below) overcomes these issues. As
described in more detail below, we measure differences in support for various policies unrelated to
militancy between two experimental groupsthose told only about the policy and those told a
militant organization supports the policy. The difference between the two conditions reveals how
much policy support increases or decreases as a consequence of being associated with a militant
group, and therefore is an indirect measure of support for the group. Unlike a direct measure,
however, non-response and social desirability are less prominent since respondents are reacting to
the policy and not to the group itself. By asking respondents about multiple policy issues and

Work that suffers from these issues includes Shapiro and Fair (2010); Fair, Ramsay, and Kull (2008);
WorldPublicOpinion.org (February 2009); the Pew Global Attitudes Surveys (2002, 2004, 2005, 2006, 2007,
and 2008); and WorldPublicOpinion.org (July 2009).

randomizing the pairing of issue with group, we can identify both average attitudes towards
militancy across groups as well as support for individual groups in ways that are unlikely to be biased
by the details of any specific policy.
Using this approach, we find that support for a set of liberal democratic valuesproperty
rights, free speech, independent courts, the ability of citizens to elect representatives, a separation of
civilian and military power, and freedom of assemblyincreases support for militancy. Moving
across the range of this index measuring support for these values is associated with a 4-5 percentage
point increase in support for militant groups that have engaged in violent political activity both
within and outside of Pakistan. Consistent with the principle of azadi (freedom or self-determination
in Urdu), this result is driven by those who believe that Muslim rights and sovereignty are being
violated in Kashmir and in Afghanistan. In other words, supporters of democratic rights are more
likely to favor militant groups if they believe that those militants are fighting against foreign forces
that are denying Muslims their rights.
The remainder of the paper is organized as follows. The first section provides a brief
background on militant groups in Pakistan and the political claims they make. Next, we provide a
theoretical basis for our expected hypotheses, centered around the concept of azadi. We then
describe our survey in detail including how we measure the core dependent and independent
variables. Subsequently, we present our methods of analysis. We conclude by presenting the results
and discussing their implications for the study of political violence as well as for foreign policy.

Militancy in Pakistan
Pakistan has employed Islamist militancy in India and Afghanistan as a tool of foreign policy
since 1947 (the year Pakistan became independent) and this continues to date (Swami 2007;,;
Hussain 2005, Jamal 2009). Many of these groups have also conducted attacks on Pakistani soil,
3

both against the state and against civilians they oppose on political or religious grounds. The militant
landscape in Pakistan today is populated by groups that vary in their sectarian commitments,
targeting choices, theatre of operations, ethnicity of operatives, and political objectives. To
understand how views on democracy might relate to popular support for these groups, a nuanced
picture of Pakistani militant organizations is in order. This section therefore outlines the major
groups asked about in our survey.

Militants Fighting in Kashmir

There are several organizations Pakistanis group under the title of Kashmiri tanzeems
(Kashmiri groups). Jaish-e-Mohammad (JM), Harkat-ul-Ansar/Harkat-ul-Mujahideen (HUA/HUM),
and their splinter groups have traditionally focused upon Kashmir and while they recruit within
Pakistan proper, their recruitment materials suggest a Kashmir-oriented mission. In recent years JM
has become intimately involved with the Pakistan Taliban and has provided suicide attackers for
assaults on Pakistani targets and international targets within Pakistan. A separate set of groups
operating in Kashmir includes Hizbol Mujahideen, al Badr, and related factions. These groups
primarily recruit Kashmiris and generally have remained focused on securing autonomy or
independence for Kashmir. They have not been involved with the Pakistan Taliban, and have not
targeted the Pakistani state or international targets within Pakistan.
The most prominent of Kashmiri tanzeem is Lashkar-e-Taiba (LeT), which has operated in
Indian-administered Kashmir for much of the 1990s (Abou Zahab 2007). LeT conducted its first
attack outside of Kashmir in 2000 and in recent years has attacked international targets in Indiathe
November 2008 Mumbai hotel attacks are the most prominent exampleas well as U.S. and allied
forces fighting in Afghanistan (Fair 2010). LeT has not targeted the Pakistani state nor has it
pursued western targets within Pakistan.

Afghan Taliban
4

As is well known, the Taliban government achieved dominance over most of Afghanistan in
1996 with Pakistani assistance (Rubin 2002). The September 11, 2001, terrorist attacks made it
impossible for Islamabad to continue supporting the Taliban (Musharraf 2006) and when the United
States-led coalition routed the Taliban in late-2001 many fled to Pakistans tribal areas to regroup. In
2005, the Afghan Taliban launched a renewed insurgent campaign run by leadership shuras in Quetta,
Peshawar, and Karachi (Levin 2009). The Afghan Taliban, despite considerable organizational
changes since 2001, remain focused on ousting foreign forces, aiding workers and other foreign
civilians from Afghanistan, overthrowing the Karzai regime, and restoring their role in governing
Afghanistan (Giustozzi 2009).

Pakistan Taliban
Since 2004 a cluster of militant groups whose activists describe themselves as Pakistani
Taliban have developed in Pakistan.2 While we were unable to measure support for these groups
due to the high level of political sensitivity surrounding them when our survey was fielded,
understanding the differences between them and the Afghan Taliban is important for interpreting
our results (Fair 2011).
The goals of the militants grouped by Pakistanis as the Pakistan Taliban are focused on
undermining the Pakistani state in select areas and establishing their own parallel governance
structures organized around commanders particular understanding of Sharia. At the time our
survey was in the field these groups had conducted few operations outside of attacking police forces
in the FATA and parts of the Khyber Paktunkhwa (KPK, formerly the Northwest Frontier Province
2

The Pakistan Taliban became prominent in early 2006 as local militias established micro-emirates in large
swathes of Pakistans Pashtun areas. Popular characterization of all Pakistan Taliban as being part of the
Tehreek-e-Taliban Pakistan (TTP) are incorrect, the term most accurately refers to a loose group of local
militias espousing a particular view of sharia law. The so-called Talibanization of the tribal areas began in
North and South Waziristan, but quickly spread to parts of the other tribal agencies. Beginning in 2007
local Taliban also emerged in parts of KPK (previously known as Northwest Frontier Province or
NWFP).The Pakistan army has engaged in various operations to contend with these militant groups (Jones
and Fair 2010).

or NWFP). This unfortunately changed in subsequent months as TTP-affiliated militants conducted

attacks across Pakistan killing thousands in response to government offensives against them.

Al Qaida
The most important militant group operating in Pakistan to Western policy makers and politicians is
al-Qaida, the group responsible for the September 11, 2001 attacks. Former British Prime Minister
Gordon Brown summed up these concerns when he reported that three quarters of the most
serious plots investigated by the British authorities have links to al Qaida in Pakistan. (Coates and
Page 2008). Important al Qaida leaders remain in the FATA and many al Qaida operativesAbu
Zubaidah, Khalid Sheikh Mohammad, and othershave been arrested in Pakistani cities. Al-Qaida
operatives in Pakistan have targeted the Pakistani state and executed terrorist plots targeting the
West and its allies. The July 7, 2005, bombings in London have been linked to al-Qaida in Pakistan,
for example, as have at least five foiled plots since 2004 (Jones and Fair, 2010).3

Sectarian Tanzeems
Pakistan is also home to a number of militant groups seeking to advance a sectarian agenda.
These firqavarana tanzeems include the anti-Shia Lashkar-e-Jhangvi (LeJ) and Sipah-e-Sahaba Pakistan
(SSP).4 The Sunni sectarian groups grew to prominence in the 1980s and are now a well-established
part of Pakistans political landscape (Nasr 2000a). In the past, Shia sectarian groups targeted Sunni
Muslims, although these groups have largely disappeared.
The anti-Shia groups all claim to be fighting for a Sunni Deobandi Pakistan by purging the
country of Shias, whom they view as apostates.5 Their actions typically take the form of attacks on
3

Many Pakistanis are dubious about the existence of al-Qaida per se. All focus group participants in our
pre-testing, however, understood what we were referring to when we explained that al-Qaida was Osama
bin Ladens militia.
4
Many of these groups have been proscribed numerous times only to re-emerge. Many now operate under
new names. We use the names which are likely to be most familiar to readers.
5
While an exact accounting of Shia in Pakistan is impossible because the Pakistani census is not fielded in
areas where Shia are populous (e.g. the Northern Areas), they are believed to comprise 20 percent of the
population (CIA 2009).

Shiite mosques and community gatherings and they have periodically attacked Christian and
Ahmediya targets as well. In reality, a great deal of the anti-Shia violence is motivated by class issues
and urbanization. The large land-holding families in Pakistan have historically been Shia and have
not treated their tenant farmers well. Thus a class agenda has been executed through a narrative of
apostasy (Nasr 2000b; Zaman 1988).

Consequences of Militant Violence in Pakistan

Militant violence has long been a fact of life in Pakistan. Since the earliest years of the state
there have been ethno-nationalist insurgencies ongoing in Balochistan, Sindh, and KPK, some of
which continue to this day at low levels.6 Sectarian tanzeems (e.g. LeJ and SSP), which were involved
in the anti-Soviet Jihad and which are now assisting the Afghan and Pakistani Taliban, have been
conducting attacks since 1979 in key districts and cities in Southern Punjab as well as in the major
provincial capitals (e.g. Quetta, Lahore, Peshawar and Karachi). In the last five years, these groups
and some of the Kashmir tanzeems (i.e. Jaish-e-Mohammad) have been targeting Pakistani security
forces, civilian government figures and civilian targets in FATA and adjacent territories under the
umbrella of the Pakistan Taliban. They have conducted attacks throughout KPK (especially
Peshawar and its environs), and hit Islamabad, Lahore, and Karachi (Gul 2009; Jones and Fair 2010).
In the year before our survey was fielded, there 11,429 people killed or injured in terrorist attacks in
KPK, 3,788 in Balochistan, 4,424 in Punjab, and 1,791 in Sindh.7

Theoretical Overview: The Concept of Azadi

Understanding the concept of azadi, which many Islamist militant groups claim to be
fighting for, is critical for making sense of the politics of militancy in Pakistan. Loosely translated,
6

A successful ethno-nationalist insurgency in Bengali-dominated East Pakistan led to the creation of

Bangladesh after the 1971 Indo-Pakistan war
7
Author calculations based on data from the National Counter Terrorism Centers Worldwide Incident
Tracking System.

azadi refers to the combination of freedom and self-determination at the level of a polity (e.g.
especially Afghans or Kashmiris). The concept is redolent of, but not isomorphic with, what we
might term democracy. Azadi fundamentally conveys a sense that politics should be organized by
and answerable to the groups seeking freedom rather than by the government or military forces (be
they foreign or domestic) that control these populations by force. Assessing how azadi has been
used over the years to justify militants actions suggests several testable hypotheses about the
relationship between support for democratic values and support for militant groups in the Pakistani
context.
Azadi literally means freedom in Urdu (as well as Hindi, Dari, Persian, Pashto among other
related languages), with explicit reference to political self-determination of a specific group of
people. During the period of British colonization, azadi referred to freedom from British
occupation and an assertion of Indian self-rule where Indian referenced the indigenous population
within the territorial dominion of the British Raj. After partition of the sub-continent into India and
Pakistan in 1947, the concept of azadi has been used by a variety of separatist groups to assert subnational autonomy and/or freedom in both post-partition India and Pakistan. For this paper, the key
expression of sub-national azadi is that relating to the disputed territory of Kashmir.
One of the principal means for the Pakistani state in this dispute has been using Islamist
militants as proxies to conduct attacks in Indian Kashmir and elsewhere (Fair 2011). The mobilizing
narrative for these groups has been attaining azadi or freedom for Muslim Kashmiris living under
Indian (e.g. Hindu) oppression.8 Of course, azadi in most of this constructed discourse implies
8

This is an oversimplification of course. Residents of the disputed area of Jammu and Kashmir under
Indian control include Hindus, Muslims, Sikhs, and Budhists among others. Since violence there erupted in
the late 1980s, there has been considerable ethnic cleansing with Hindus moving out of the valley of
Kashmir to Jammu along with Sikhs. Buddhists have tended to be in the Leh-Ladakh area and remain so.
Currently, the dispute is most intense over the valley of Kashmir, which is dominated by Muslims. The
Pakistani claims that India is a Hindu state is also deeply problematic because India is technically a
democratic state that, while not secular in the American sense, adheres to a notion of religious equality.
Nonetheless, due to the preponderance of Hindus in the security forces, this facile and polemic
characterization of Hindu oppression is sustainable for many Pakistanis.

that if given the choice, Indias Kashmiris would join Muslim Pakistan to free themselves from
Hindu domination.9 Reflective of this discourse, Pakistan-administered Kashmir is called Azad10
Kashmir (Free Kashmir) while that under Indias administration is called Maqbuza Kashmir (or
Occupied Kashmir).
Kashmiri azadi is crucially important in Pakistani domestic politics and society. Pakistans
textbooks rehearse the elemental unfairness of the partition which deprived Pakistan of Kashmir.
Pakistans media on a daily basis report on the Indian states abuses and other missteps in Kashmir.
Pakistans civilian and military leaders alike refer to the militant groups which claim to fight on
behalf of Kashmirs freedom as freedom fighters rather than terrorists. On February 5, Pakistan
celebrates Kashmir Day with demonstrations in Azad Kashmir and elsewhere to show solidarity
with Kashmiris living under Indian occupation. Pakistanis driving to Azad Kashmir do so on the
Srinagar Highway, named for the capital of Maqbuza Kashmir. Pakistans cities are strewn with
public commemorations and memorializations of Kashmir and many of Pakistans leadership (e.g.
the Sharifs of the Pakistan Muslim League) are Kashmiri. Many products such as cooking oil are
sold and marketed under the brand name Kashmir. Pakistans various tanzeems operating in
Kashmir and the rest of India do so under the stated aims of freeing Indias Muslims from Hindu
hegemony and allowing them to join Pakistan (Fair 2011).
The notion of azadi also applies to Afghanistan in Pakistani discourse, albeit in a much less
intense fashion. In the 1980s, mujahideen were mobilized to free the Afghans from the secular
Soviet occupation. Throughout the 1990s, Pakistan continued to justify its involvement and that of
its so-called mujahideen proxies in Afghanistan by arguing that it was liberating Afghanistan
from vicious warlords enjoying the support of India, Russia and Iran among others (Rashid, 2000).
9

Recent polling results published the Chatham House finds evidence that, in fact, Kashmiris on both sides
of the LOC prefer independence from both India and Pakistan (Brandnock 2010). However, few Pakistanis
believe this to be the case (WorldPublicOpinion 2008).
10
Azadi is the noun related to the adjective azad.

With the U.S. invasion in 2001, Pakistanis have again viewed Afghanistan as occupied and the
Afghan Taliban as a legitimate group fighting jihad for the azadi of Afghans. Hence, concepts of
democracy, self-determination, and violent uprising are intertwined in Pakistani culture.
This history suggests two testable hypotheses:
Hypothesis 1 (H1): Pakistanis more supportive of liberal democratic principles consistent with azadi
should be more supportive of militant groups operating from Pakistan.
Hypothesis 2 (H2): The relationship posited by H1 should be strongest among Pakistanis whose beliefs
about Muslims in Kashmir and Indias presence in the region map strongly onto the azadi narrative.

The Survey
Our survey was designed to achieve three goals. First, we wanted to survey a representative
sample of the Pakistani population, including rural and urban areas in each of Pakistans four main
provinces. Second, we sought to measure attitudes towards specific militant organizations in a way
that minimized item non-response on sensitive questions which had plagued previous surveys in
Pakistan.11 Third, we wanted to mitigate social desirability bias in measuring affect towards militants,
our key dependent variable. As is well known, respondents in many survey settings anticipate the
views of the enumerator and thus answer in ways to please or seem high-status to the enumerator
(Krosnick 1999; Marlowe and Crowne 1964, p. 109). These tendencies may be exacerbated on
sensitive issues where fear and the desire to avoid embarrassment come into play. In Pakistan,
respondents can often determine significant information about class, ethnicity, and sectarian
orientation based on the name and accent of the enumerators. This makes social desirability
concerns even stronger for surveys studying the politics of militancy in Pakistan, as respondents may
11

Surveys in Pakistan which ask directly about affect towards militant groups obtain dont know/no
opinion rates in the range of 40% (Terror Free Tomorrow 2008; Pew 2009). Surveys which indirectly
measure affect by asking whether groups operating in Pakistan are a problem (IRI 2009) or pose a
threat to the vital interests of Pakistan (WPO 2009) still obtain item non-response rates as high as 31
percent.

10

be wary to signal pro-militant views to high-status enumerators.

Working with our Pakistani partners, Socio-Economic Development Consultants (SEDCO),
we drew a random sample of 6,000 adult Pakistani men and women from the four normal
provinces of the country which are governed by Pakistans 1973 constitution: Punjab, Sindh, KPK,
and Balochistan using the Pakistan Federal Bureau of Statistics sample frame. The respondents were
selected randomly within 500 primary sampling units (PSU), 332 in rural areas and 168 in urban ones
(following the rural/urban breakdown in the Pakistan census). We substantially oversampled in the
smaller provinces (Balochistan and KPK) to ensure we could generate valid estimates in these
sparsely populated provinces. We calculated post-stratification survey weights based on population
figures from the 1998 census, the most recent one available. Following procedures outlined by Lee
and Forthofer (2006), all analyses reported below were weighted and clustered to account for survey
design effects.
The face-to-face questionnaire was fielded by six mixed-gender teams between April 21,
2009 and May 25, 2009. The overall response rate was over 90 percent, which rivals the extremely
high response rates achieved by the United States Census Bureau. Online Appendix A reports the
sample demographics. Full question wordings are provided in Online Appendix B. All variables
described below were coded to lie between 0 and 1, so that we can easily interpret a regression
coefficient as representing a % change in the dependent variable associated with moving from the
lowest possible value to the highest possible value of the independent variable.
We pre-tested the questionnaire to residents of Islamabad, Peshawar, and Rawalpindi
between March 20 and 26, 2009, in order to assess the functioning of the items and experiments.
Several design decisions came about as a result of what we learned during pretesting.

Measuring Support for Islamist Militant Organizations: The Endorsement Experiment

11

Asking respondents directly whether they support militant organizations has numerous
problems in places suffering from political violence. First, and perhaps most importantly, it is unsafe
for enumerators and respondents to discuss such issues. Second, item non-response rates to such
sensitive questions are often quite high given that respondents fear that providing the wrong
answer will threaten their own and their familys safety. We therefore use an endorsement
experiment to measure support for specific Islamist militant organizations.12
The experiment involves assessing support for various real policies, which are relatively well
known but about which Pakistanis do not have strong feelings (as we learned during pretesting) and
works as follows:
-

Respondents are randomly assigned to treatment or control groups (one-half of the sample
is assigned to each group).

Respondents in the control group were asked their level of support for four policies, which
is measured on a five-point scale, recoded to lie between 0 and 1 for the analysis.

Respondents in the treatment group are asked identical questions but then are told that one
of four groups mentioned in the first section supports the policy in question. Which group is
associated with each of the four policies is randomized within the treatment group.

The difference in means between treatment and control groups provides a measure of affect
towards the groups, since the only difference between the treatment and control conditions
is the group endorsement.

Figure 1 provides a sample question, showing the treatment and control questions, and illustrates the
randomization procedure in visual form. Online Appendix B describes all questions measuring
support. Online Appendix C presents randomization checks showing balance on observables
between treatment and control groups.
12

This approach builds on the technique introduced in Lupia and McCubbins (1998).

12

The advantage of this approach is that the militant organization is not the primary object of
evaluation; the policy is. We expected respondents to be more willing to share their opinions on
uncontroversial policies rather than controversial groups. However, by embedding endorsements
within the questions, we are able to indirectly ascertain support for militant organizations. Because
we randomize both assignment to the group endorsement and the pairing of issues with groups, any
difference in policy support can be attributed solely to the group.
We used this method to measure support for four groups: the Kashmiri tanzeems, the
Afghan Taliban, al-Qaida, and the sectarian tanzeems.13 This required asking about four policy
issues: polio vaccinations, reforming the frontier crimes regulation (the legal code governing the
FATA), redefining the Durand line (the border separating Pakistan from Afghanistan), and requiring
madrassas to teach math and science. By randomizing which group is associated with which policy
among the treatment group, we control for order effects and randomize the pairing of issue with
group. This allows us to identify effects for multiple groups that are unlikely to be biased by the
details of any specific policy.
For an endorsement experiment of this type to work the policies need to have two
characteristics. First, they need to be ones about which respondents do not have overly strong prior
opinions so that a groups endorsement might affect their evaluation of the policy. This procedure
would not work in the U.S., for example, if one asked about banning abortion, for which prior
attitudes are strong. Second, the policies have to be at least somewhat familiar to respondents since
the group endorsement has to be meaningful and salient. For example, in the U.S., one could not
ask about an obscure mining regulation since respondents may not provide meaningful responses
13

We did not employ this method to assess support for the Pakistani Taliban. Within our budget for the
survey we could only interview 6,000 respondents (twice as large as any other extent survey of Pakistani
public opinion). This meant we could only study four groups (i.e., divide the sample into four cells) while
getting reasonable precision at the provincial level. Given this constraint, we omitted an endorsement
experiment on the Pakistan Taliban because: (1) at the time the survey was designed, the group was not as
prominent as it has since become; and (2) there were safety concerns for enumerators as mentioned above.

13

and endorsements may have limited impacts. While the policies we studied may seem high valence
to professional students of politics, they do not appear to be so for most Pakistanis. During pretesting, we found that most respondents knew about all four issues but did not have strong opinions
on them. Our enumerators likewise felt these issues would be ones respondents would know
something about but at the same time not have extremely rigid positions, a telling fact since our
enumerators were all professionals averaging 4.6 years of experience.
There is empirical evidence in the survey that attests to the validity of the policies as well.
Figure 2 shows the distribution of attitudes of policies in the control group. Importantly, it is single
peaked on all but one issue, implying respondents do not have highly-polarized attitudes, as they
would for something like abortion in the United States. The variance in responses to these policies
in the control group was fairly large, ranging from .98 (polio vaccinations) to 1.28 (redefining the
Durand line) on a five-point scale, suggesting potential malleability. For comparison purposes the
1987 General Social Survey (GSS) asked whether respondents agreed that The government should
provide a decent standard of living for the unemployed on a five-point scale. The variance in
responses in the GSS was approximately .657.
This approach unambiguously drove down item non-response. Our survey posed a number
of direct questions (i.e., without an endorsement experiment) such as What is the effect group Xs
actions on their cause? Non-response on these items ranged from 22% for al-Qaida to 6% for the
Kashmir Tanzeem. Item non-response on the endorsement experiment questions, by contrast,
ranged from a high of 7.6% for al-Qaida endorsing Frontier Crimes Regulation reform to a meager
0.6% for the sectarian tanzeems endorsing polio vaccinations. While this approach is not perfect, the
low item non-response rate in our survey provides prima facie evidence that this technique reduced
respondents concerns about reporting sensitive information.14 That the endorsement experiment
14

Compared to other surveys, the contrast between direct questions and this approach is even starker. The
PIPA 2007 survey of urban Pakistanis, for example, had a DK/NR rate of around 20 percent on most of

14

drives down item non-response is not necessarily evidence that it also ameliorates social desirability
bias. Nonetheless, one would need to tell a fairly contorted story to explain why a technique that
drives down item non-response so dramatically would fail to address social desirability biases that
stems from respondents concerns about how enumerators will react to their answers.
To construct our dependent variable of support for militancy, we measure the average
support the respondent reports for the four policies. Recall that one of the four militant groups was
randomly assigned to be associated with each policy in the treatment group. Below, we leverage
random assignment into treatment (endorsement) and control to measure differential support for
militancyas proxied by support for the policies. The main dependent variable therefore was a
twenty-point scale, recoded to lie between 0 (no support for all four policies) to 1 (a great deal of
support for all four policies). In the control group, the policy scale had a mean value of .79 (s.d. = .
15). As described below, we also examined support for each of the groups individually.

Measuring Support for Democratic Values

We measured support for six core features of liberal democratic societies using questions which
draw on the widely-used Freedom in the World (FIW) survey (Freedom House 2011).
Free Speech. How important is it that individuals be able to express their political views, even
though other people may not agree with them? (Freedom of Expression and Beliefs module)
Independent Courts. How important is it for you to live in a country where the decisions of the
courts are independent from influence by political and military authorities? (Rule of Law module)
the questions but for questions about the activities of Pakistan-based militant groups, the DK/NR rate was
sometimes in excess of 50 percent. When PIPA asked different samples of Pakistanis How do you feel
about al Qaeda? in 2007, 2008 and 2009, DK/NR rates were 68 percent, 47 percent and 13 percent,
respectively. When Pakistanis were asked who perpetrated the 9/11 attacks, DK/NR rates were 63 percent
and 72 percent in 2007 and 2008, respectively (Fair et al. 2008). The Pew Global Attitudes Survey
encountered similar problems when they asked (predominantly urban) Pakistanis whether they have a very
favorable, somewhat favorable, somewhat unfavorable, or very unfavorable opinion of al Qaida. In 2008
and 2009, the DK/NR rates were 41 percent and 30 percent, respectively. When the same question was
posed about the Taliban in 2008 and 2009, the DK/NR rates were 40 percent and 20 percent, respectively
(Pew 2009).

15

Freedom of Assembly. How important is it that individuals be able to meet with others to work on
political issues? (Associational and Organizational Rights module)
Being Governed by Elected Representatives. How important is it for you to live in a country that is
governed by representatives elected by the people? (Functioning of Government module)
Property Rights. How important is it that individual property rights be secure? This means the
state cannot take away their things without proper court proceedings. (Personal Autonomy and
Individual Rights module)
Having Civilian Control over the Military. The 1973 Constitution of Pakistan says civilians should
control the military. This means the military cannot take action without orders from civilian leaders.
In your opinion, how much control should civilians have over the military? (Functioning of
Government module)
The first five items were measured on a five-point scale (extremely important, very
important, moderately important, slightly important, not important at all). The civilian
control item was measured on a different five-point scale (complete control, a lot of control, a
moderate amount of control, a little control, no control at all).
As shown in Table 1, about half of respondents select the most-democratic response
(extremely important or complete control) and very few select response options in the bottom
categories expressing little support for democracy. Accordingly, we bifurcate respondents into two
groupsthose selecting the highest response category and all others. We also estimated
specifications in which responses were treated as continuous measures and we obtained similar
results. In addition, we construct a scale in which we average the six items together to reduce
measurement error. Cronbachs alpha for the six items was .75, suggesting a high level of scale
reliability. The democratic support index had a mean of .48 (s.d. = .33).
To test Hypothesis 2, we also needed to measure respondents beliefs about the status of
16

Muslims living in Kashmir. To do so, we constructed a three-point scale measuring perceptions that
Muslims are being oppressed in Kashmir based on two binary indicators.15 The first question asked
respondents How well does India protect the rights of its Muslim citizens in Kashmir? (response
options: extremely well, somewhat well, neither well nor poorly, somewhat poorly,
extremely poorly.16 Respondents answering extremely well and somewhat well were coded as
0 and all others were coded as 1. The second question asked respondents Thinking about the
political preferences of Muslims in occupied Kashmir, please tell us which statement you agree with
the most (response options: In occupied Kashmir, the majority of Muslims want to be part of
India, In occupied Kashmir, the majority of Muslims want an independent state, In occupied
Kashmir, the majority of Muslims want to be part of Pakistan). Respondents answering that
Muslims want to be part of India were coded as 0 and all others were coded as 1.

Control Variables
We additionally measured several control variables, which we include in our models both
additively and multiplicatively: gender, marital status, age, access to the Internet, ability to read,
write, and do math, education level, income, and sectarian affiliation (Sunni/Shia). These variables
have all been cited as potential correlates of support for violent politics including: age (Russel and
Miller, 1977), marriage (Berrebi, 2007), media access (Bell, 1978; Dowling 2006), education (Becker,
1968), income (Muller, 1985), and religion (Juergensmeyer, 2003). We also controlled for various
attitudinal measures including views on the U.S. governments influence on the world, views on the
U.S. governments influence on Pakistan, and belief that Sharia law is about physical punishment.

15

We also examined the conditional effects of these two variables in isolation and obtained similar results
to using the averaged measure.
16
Prior to asking this question, we randomly presented some respondents with information about the
relative strength of the Indian and Pakistani militaries. This manipulation had no significant or substantial
effect of responses to this question.

17

We hypothesize that negative views of the U.S. and belief in the corporal punishment aspects of
Sharia should be positively related to militancy. Finally, in the regression models, we also include
province fixed effects.

Methods of Analysis
To test Hypothesis 1, we estimate the following OLS regression model:
Pi = + 1Ti + 2Di + 3(Ti Di) + i + i (1)
where Pi is a continuous variable representing support for the four target policies, Ti is a dummy
variable representing assignment to the group endorsement group, Di is a continuous variable
ranging from support for zero democratic values (0) to support for all six values (1), i is a vector of
region fixed effects, and i is a normally-distributed error term. 1 represents our measure of support
for militancythe change in support for the policy due to the group endorsementamong
respondents who score lowest on the democracy index. 2 represents the effect of democratic values
on support for policies among respondents in the control group. 1 + 3 represents support for
militancy among respondents who are the strongest supporters of democracy. Hence, the key
parameter of interest is 3, from which we can derive the marginal effect of support for democracy
on support for militancy (following Brambor et al. 2005).
Note that the difference in variances across policies suggests that some may exhibit greater
treatment effects than others because prior attitudes are less well-formed. We therefore use the
variance of the responses in the control group to proxy looseness of pre-treatment attitudes and
account for its influence by weighting each policy response by this variance. Hence, we place greater
weight on policies where the survey responses lead us to expect a greater likelihood that attitudes

18

will be shifted in response to the endorsements.17

To test the robustness of our results, we also estimate a series of more-saturated models, the
most complex of which is represented by equation (2):
Pi = + 1Ti + 2Di + 3(Ti Di) + i + xi + zi + Tixi + Tizi + i (2)
where xi represents a vector of demographic control variables and zi represents a vector of attitudinal
control variables. Note that equation (2) includes not only the main effects of these controls on
support for the policy but also the interactive effects with the treatment dummy.
To test Hypothesis 2, we estimate an analogous set of models:
Pi = + 1Ti + 2Di + 3Ki + 3Ki + 4(Ti Di) + 5(Ti Ki) + 6(Di Ki) + 7(Ti Di Ki) + i + i (3)
Pi = + 1Ti + 2Di + 3Ki + 4(Ti Di) + 5(Ti Ki) + 6(Di Ki) + 7(Ti Di Ki)
+ i + xi + zi + Tixi + Tizi + i (4)

where Ki represents respondents beliefs about the state of Muslims in Kashmir. As mentioned in the
section entitled Meaning Support for Democratic Values, it is measured using the index based on
three questions about each groups goals. Interpreting these models is complex; we follow
procedures laid out by Brambor et al. (2005). The main parameter of interest is represented by 7,
which allows us to test whether the democracy-militancy relationship implied by H1 is significantly
stronger amongst respondents high on the Kashmir beliefs index than among those lower on the
index.

Results
Consistent with Hypothesis 1, we find that support for democratic values increases support
17

The results are substantively similar without this weighting and so we report weighted results
throughout as we believe they more accurately capture the impact of cues on attitudes. The weight vector
w for the four policies (vaccination plan, FCR reforms, Durand line, curriculum reform) was: (.983, 1.15,
1.28, 1.18), meaning that the weight for the control group was the average of these four individual weights
(1.15). The post-stratification weight was multiplied by w to produce the overall sampling weight.

19

for militancy across all groups. In the first column of Table 2, we present the estimates from the
simple model described in equation (1). Controlling for provincial differences in support, we find
that among those scoring zero on the democracy scale, the group endorsement actually decreases
support for the policies by about 2.8 percentage points (1 = -.028, p < .01, two-tailed). However,
among the strongest supporters of democracy, we estimate the treatment effect of the endorsements
to be positive 2.3 percentage points (1 +3 = .023, p = .04). Therefore, the overall effect of
democracy on support for militancy is 4.5 percentage points (3 = .045, p < .01). In Figure 3, we plot
the marginal effect of support for democracy along with the associated 95% confidence interval.
Confirming the model estimates, among low supporters of democracy, the treatment effect of the
endorsement cues is negative. However, strong supports of democracy are more supportive of the
policies as a result of the endorsements.
[FIGURE 3 ABOUT HERE]
How big is this effect in substantive terms? In the control group, support for democratic
values increases support for the government policies (as one would expect) by about 12.2 percentage
points, as indicated by the parameter estimate of 2. Hence, our difference-in-difference estimate
represents about 37% of this main effect, and is therefore substantively meaningful. Another way to
assess the effect size is to compare it to the effect of incomean expectedly strong predictoron
support for the policies. Unsurprisingly, going from the bottom income group to the top income
group decreases support for these social services by 6.4 percentage points. Hence, the difference-indifference estimate (3) represents 70% of the main income effect.
This finding is highly robust. In column two of Table 2, we present estimates from a
regression specification including demographic controls, along with a dummy variable for
respondents who did not answer the income question. In column three, we listwise delete cases for

20

which we do not have a valid income response. In column four, we include attitudinal controls in
the model. Finally, in column five, we estimate the model represented by equation (2), which
includes all the main and interactive effects. Our estimate of 3 is highly stable across all
specifications, representing between 3.7-4.5 percent of the range of the dependent variable.
Consistent with Hypothesis 2, we find that the positive democracy-militancy relationship
shown in Figure 2 is driven by those who feel that the groups are fighting for democratic values. We
present estimates from equation (3) in the first column of Table 3. The parameter estimate of 7 is
positive and statistically significant (7 = .127, p = .08), indicating that the effect of democracy on
militancy increases by 12.7% as we move from belief that Muslims are not being mistreated in
Kashmir and desire to live under Indian control (Ki = 0) to belief that Muslims are disenfranchised
(Ki = 1). Figure 4 illustrates these results. Note that the slope of the relationship between democracy
and militancy is essentially flat among those low on the Kashmir index (Ki < 1), and becomes
positive and steep as the value of the index increases to 1. As shown in columns (2)-(5) of Table 3,
this result only becomes stronger as additional variables are added to the model and the overall
estimate of 7 is stable.
Finally, we analyzed the results separately for each militant group and each democratic value.
As shown in Table 4, the estimate of 3 from equation (1) is positive and significant for all four
groups. A shown in Table 5, the estimate of 3 is correctly signed for all six components of the
democracy index, with the strongest relationship being for four particular indicatorsproperty
rights, independent courts, elected representatives, and freedom of assembly.

Discussion
To better understand the politics of militancy in Pakistan and to shed light on larger theories
21

about the relationship between democratic values and support for violent political organizations, we
designed and conducted a 6,000-person nationally and provincially representative survey of Pakistani
adults, measuring affect towards four specific militant organizations. We applied a novel
measurement strategy to mitigate social desirability bias and item non-response given the sensitive
nature of militancy in the region. Our endorsement experiment overcomes several issues that have
plagued past efforts to use surveys to study the politics of militancy.
Using this innovative approach we find that support for a set of core liberal democratic
values correlates with higher support for militant groups. This finding contravenes the conventional
wisdom which underlies recent U.S. policy approaches to Pakistan in particular and the Muslim
World in general. We measure support for democratic values using an index that aggregates support
for six key values: property rights, free speech, independent courts, being ruled by elected
representatives, civilian control of the military, and freedom of assembly. Moving from the lowest
value on this index to the top is associated with a 4-5 percentage point increase in support for
militant groups.
This result may seem puzzling, but it makes sense in the particular context of Pakistan where
militant groups (and their advocates in government) have long justified their actions as defending
azadi, a concept that loosely translates as freedom and self-determination. We find exactly what one
would expect if this history were driving our results; the democracy-militancy relationship is
strongest among respondents whose beliefs about Kashmir are consistent with the azadi narrative
and who believe the groups are fighting for justice, democracy, and to protect Muslims.
Moving beyond Pakistan, our larger theoretical contribution is to demonstrate that the
relationships between individual values on the one hand, and attitudes towards political
organizations on the other, depends strongly on beliefs about the political context. Accordingly,
there is no simple mapping between personal attitudes that seem normatively attractive, such as a
22

belief in individual liberty, and rejection of normatively unappealing methods of political

contestation. In Pakistan, for example, some militant groups rhetoric justifying fighting for azadi has
been so widely accepted that it is exactly those who believe most deeply in democratic values that
were most supportive of violent groups.
The policy implications of this research are stark. Whether democratic values are a force for
peace or for conflict depends on the how people understand the political context. Those seeking to
promote pacific dispute resolution and orderly politics in Pakistan and elsewhere therefore need to
move beyond efforts to delegitimize violence in a normative sense to efforts to convince potential
supporters that it is a counterproductive method. In less fraught political realms advocates should
consider that it may be easier to convince people that the facts of the world imply different political
behavior than it is to change their underlying attitudes.
Finally, this paper suggests some new lines of inquiry for students of both violent politics
and political behavior. For those studying violent politics, the paper highlights the potential
importance of learning about how beliefs about the strategic environment interact with deeplyseated attitudes to generate support for specific militant organizations. Future studies can potentially
manipulate beliefs about the strategic impact of the groups actions and measure resulting changes in
support for militancy. For those studying political behavior, the paper shows clearly that beliefs
about the political environment interact with long-standing beliefs to generate attitudes towards
specific actors. This means one cannot look solely at what is going on inside peoples heads, but also
how those attitudes interact with political structures, organizations, and institutions.

23

Table 1: Distribution of Democratic Values Variables

Elected Representatives
Extremely Important
Very Important
Moderately Important
Slightly Important
Not Important at All
N

49.7%
29.8
14.2
4.0
2.4
5891

Property Rights
Extremely Important
Very Important
Moderately Important
Slightly Important
Not Important at All
N

67.1%
18.0
9.6
3.9
1.4
5914

Independent Courts
Extremely Important
Very Important
Moderately Important
Slightly Important
Not Important at All
N

54.9%
25.9
14.4
3.2
1.6
5885

Civilian/Military Separation
Extremely Important
Very Important
Moderately Important
Slightly Important
Not Important at All
N

39.6%
17.6
25.7
10.7
6.4
5770

Free Speech
Extremely Important
Very Important
Moderately Important
Slightly Important
Not Important at All
N

48.1%
33.3
13.5
4.1
1.0
5887

Freedom of Assembly
Extremely Important
Very Important
Moderately Important
Slightly Important
Not Important at All
N

53.5%
29.6
12.7
3.6
.6
5893

Note: Data weighted and adjusted for sampling design.

24

Table 2: Support for Democratic Values Predicts Support for Militancy

(1)

(2)

(3)

(4)

(5)

1: Group Cue

-.028***
(.007)

-.028***
(.007)

-.028***
(.006)

-.026***
(.006)

-.045*
(.026)

2: Support for Democratic Values

.122***
(.016)

.098***
(.014)

.096***
(.015)

.093***
(.014)

.092***
(.014)

3: Group Cue x Support for Democratic

Values

.045***
(.012)

.042***
(.011)

.043***
(.011)

.037***
(.011)

.041***
(.013)

Constant

.747***
(.011)

.844***
(.022)

.843***
(.022)

.810***
(.029)

.819***
(.031)

R2
N

.14
5358

.24
5358

.23
5183

.25
5243

.25
5243

Y
N

N
N
N

Y
Y
N
N
N
N

Y
Y
Y
N
N
N

Y
Y
N
Y
N
N

Y
Y
N
Y
Y
Y

Region Fixed Effects

Demographic Controls
Income Listwise Deleted
Attitudinal Controls
Group Cue-Demographics Interactions
Group Cue-Attitudinal Interaction

***p<.01; **p<.05; *p<.10 (two-tailed)

Note: OLS regressions predicting support for policies. Data weighted and adjusted for sampling design. Demographic controls
include: gender, martial status, age, access to Internet, possession of cellular phone, ability to read, ability to write, ability to
perform arithmetic, formal education level, income, and religion sect. Attitudinal controls include two measures of attitudes
toward United States and view of Sharia law.

25

Table 3: Beliefs about Muslims in Kashmir Moderate the Democracy-Militancy Relationship

(1)

(2)

(3)

(4)

(5)

1: Group Cue

.011
(.030)

.019
(.029)

.028
(.029)

.019
(.030)

-.002
(.037)

2: Support for Democratic Values

.183***
(.048)

.171***
(.043)

.171***
(.046)

.164***
(.042)

.165***
(.042)

3: Perception of Muslims in Kashmir

.082***
(.023)

.067***
(.022)

.076***
(.022)

.059***
(.022)

.058**
(.023)

4: Group Cue x Support for Democratic

Values

-.083
(.069)

-.103
(.064)

-.098
(.068)

-.101
(.065)

-.099
(.065)

5: Group Cue x Perception of Muslims in

Kashmir

-.042
(.033)

-.048
(.032)

-.058*
(.031)

-.045
(.032)

-.039
(.033)

6: Support for Democratic Values x

Perception of Muslims in Kashmir

-.067
(.050)

-.079
(.044)

-.080*
(.047)

-.074*
(.043)

-.076*
(.043)

7: Group Cue x Support for Democratic

Values x Perception of Muslims in Kashmir

.127*
(.072)

.144**
(.067)

.141**
(.071)

.139**
(.067)

.139**
(.068)

Constant

.675***
(.023)

.787***
(.031)

.778***
(.032)

.766***
(.036)

.775***
(.037)

R2
N

.16
5077

.25
5077

.25
4933

.26
5002

.26
5002

Y
N

N
N
N

Y
Y
N
N
N
N

Y
Y
Y
N
N
N

Y
Y
N
Y
N
N

Y
Y
N
Y
Y
Y

Region Fixed Effects

Demographic Controls
Income Listwise Deleted
Attitudinal Controls
Group Cue-Demographics Interactions
Group Cue-Attitudinal Interaction

***p<.01; **p<.05; *p<.10 (two-tailed)

Note: OLS regressions predicting support for policies. Data weighted and adjusted for sampling design. Demographic
controls include: martial status, age, access to Internet, possession of cellular phone, ability to read, ability to write, ability to
perform arithmetic, formal education level, income, and religion sect. Attitudinal controls include two measures of attitudes
toward United States and view of Sharia law.

26

Table 4: Support for Democratic Values Predicts Support for Four Militant Groups
Kashmeer
Tanzeem

Afghan
Taliban

Al Qaeda

Sectarian
Tanzeem

1: Group Cue

-.034***
(.009)

-.037***
(.009)

-.024***
(.008)

-.021**
(.008)

2: Support for Democratic Values

.091***
(.014)

.096***
(.014)

.094***
(.014)

.094***
(.014)

3: Group Cue x Support for Democratic

Values

.045***
(.015)

.043***
(.016)

.029**
(.014)

.025*
(.015)

Constant

.815***
(.031)

.801***
(.033)

.804***
(.032)

.822***
(.033)

R2
N

.16
5243

.14
5243

.15
5243

.15
5243

***p<.01; **p<.05; *p<.10 (two-tailed)

Note: OLS regressions predicting support for policies. Data weighted and adjusted for sampling design.
Regression models include region fixed effects, demographic controls, and attitudinal controls. Demographic
controls include: gender, martial status, age, access to Internet, possession of cellular phone, ability to read,
ability to write, ability to perform arithmetic, formal education level, income, and religion sect. Attitudinal
controls include two measures of attitudes toward United States and view of Sharia law.

27

Table 5: Support for Six Democratic Values Predicts Support for Militancy
Property
Rights

Free
Speech

Indep.
Courts

Elected
Reps.

Civil/Mil.
Separation

Freedom
Assembly

1: Group Cue

-.016***
(.006)

-.008
(.005)

-.022***
(.005)

-.016***
(.005)

-.011**
(.005)

-.017***
(.005)

2: Support for Democratic Values

.043***
(.008)

.028***
(.008)

.024***
(.008)

.041***
(.009)

.061***
(.008)

.019**
(.009)

3: Group Cue x Support for

Democratic Values

.013*
(.008)

.003
(.008)

.028***
(.008)

.018**
(.008)

.008
(.007)

.020**
(.008)

Constant

.819***
(.029)

.835***
(.028)

.833***
(.029)

.836***
(.028)

.814***
(.029)

.839***
(.029)

R2
N

.22
5243

.21
5243

.22
5243

.22
5243

.24
5243

.21
5243

***p<.01; **p<.05; *p<.10 (two-tailed)

Note: OLS regressions predicting support for policies. Data weighted and adjusted for sampling design. Regression models include region
fixed effects, demographic controls, and attitudinal controls. Demographic controls include: gender, martial status, age, access to Internet,
possession of cellular phone, ability to read, ability to write, ability to perform arithmetic, formal education level, income, and religion sect.
Attitudinal controls include two measures of attitudes toward United States and view of Sharia law.

28

Figure 1: Illustration of The Endorsement Experiment

Control
[POLICY DESCRIPTION]. How much do you support such a plan?
Treatment
[POLICY DESCRIPTION]. [GROUP NAME] have voiced support for this program. How
much do you support such a plan?
Randomization Procedure

29

Figure 2: Distribution of Support for Policies in Control Group

60
50
40
30
20
10
0
Vaccinations
Not at all

FCR Reforms
A Little

Peace Jirga

A Moderate Amount

A Lot

30

Curriculum Reform
A Great Deal

0.02
0.00
-0.02
-0.04

Treatment Effect of Group Cue

0.04

Figure 3: Support for Militancy by

Support for Democratic Values

0.0

0.2

0.4

0.6

0.8

Support for Democratic Values

31

1.0

Figure 4: Moderating Effect of Views of Muslims in Kashmir on Democracy-Militancy Relationship

0.05
0.00

Treatment Effect of Group Cue

-0.10

-0.05

0.05
0.00
-0.05
-0.10

Treatment Effect of Group Cue

0.10

(b) High on Kashmir Index

0.10

(a) Low on Kashmir Index

0.0

0.2

0.4

0.6

0.8

1.0

0.0

Support for Democratic Values

0.2

0.4

0.6

Support for Democratic Values

32

0.8

1.0

References
Abou Zahab, Mariam. 2002. The Regional Dimension of Sectarian Conflicts in Pakistan. Pakistan:
Nationalism Without a Nation? Ed Christophe Jaffrelot. London: Zed. 115-128.
Abou Zahab, Mariam. 2007. I Shall be Waiting at the Door of Paradise: The Pakistani Martyrs of the
Lashkar-e-Taiba (Army of the Pure). The Practice of War: Production, Reproduction and Communication of
Armed Violence. Eds. Aparna Rao et al.. New York: Berghahn Books, 133-158.
Almond, Gabriel A, and Sidney Verba. 1963. The Civic Culture; Political Attitudes and Democracy in Five Nations.
Princeton, NJ: Princeton University Press.
Becker, Gary S. 1968. Crime and Punishment: An Economic Approach. Journal of Political Economy 76 (2):
169-217.
Bell, J. Bowyer. 1978. Terrorist scripts and live-action spectaculars. Columbia Journalism Review 17 (1): 47-50.
Berrebi, Claude. 2007. Evidence about the Link Between Education, Poverty and Terrorism among
Palestinians. Peace Economics, Peace Science and Public Policy 13 (1), Article 2.
Brambor, Thomas, William Roberts Clark, and Matt Golder. 2006. Understanding Interaction Models:
Improving Empirical Analyses. Political Analysis 14 (1): 63-68.
CIA. 2009 The World Factbook -- Pakistan.
Coates, Sam, and Jeremy Page. 2008. Pakistan 'linked to 75% of all UK terror plots', warns Gordon Brown.
The Times (London), December 15, 2008.
Dalton, Russel J. 1994. Communists and Democracts: Democratic Attitudes in the Two Germanies. British
Journal of Political Science 24 (4): 469-493.
Dowling, Ralph E. 1986. Terrorism and the Media: A Rhetorical Genre. Journal of Communication 36 (1): 1224.
Fair, C. Christine. 2011. The Militant Challenge in Pakistan. Asia Policy 11 (January): 105-137.
Fair, C. Christine, Ramsay, Clay, and Kull, Steven. 2008. Pakistani Public Opinion on Democracy, Islamist Militancy,
and Relations with the US. Washington D.C.: WorldPublicOpinion.org and United States Institute of
Peace. Available at: http://www.worldpublicopinion.org/pipa/pdf/jan08/Pakistan_Jan08_rpt.pdf
(accessed December 13, 2009).
Gibson, James L. 1997. Mass Opposition to the Soviet Putsch of August 1991: Collective Action, Rational
Choice, and Democratic Values in the Former Soviet Union. American Political Science Review 91 (3):
671-684.
Giustozzi, Antonio. 2009. Koran, Kalashnikov, and Laptop: The Neo-Taliban Insurgency in Afghanistan 20022007.
New York: Columbia University Press.
Gul, Imtiaz. 2009. The Al Qaeda Connection: The Taliban and Terror in Pakistans Tribal Areas. London: Penguin.
Hamid, Shadi and Steven Brooke. 2010. Promoting Democracy to Stop Terror. Hoover Institution Policy
Review 159.

33

Huntington, Samuel P. 1984. Will More Countries Become Democratic? Political Science Quarterly 99 (2): 193218.
Husssain, Rizwan. 2005. Pakistan and the Emergence of Islamic Militancy in Afghanistan. Burlington: Ashgate.
International Republican Institute. 2009. IRI Pakistan Index. International Republican Institute. Available at:
http://www.iri.org/newsreleases/pdfs/2009%20May%2011%20IRI%20Pakistan%20Index,
%20March%207-30,%202009.pdf (accessed December 13, 2009).
Jamal, Arif. 2009. Shadow War: The Untold Story of Jihad in Kashmir. Brooklyn: Melville House.
Jones, Seth G. and Fair, C. Christine. 2010. Counterinsurgency in Pakistan. Santa Monica: RAND
Corporation.
Juergensmeyer, Mark. 2003. Terrorism in the Mind of God. London, England: University of California Press.
Kirwin, Mathew, and Wonbin Cho. 2009. Weak States and Political Violence in sub-Saharan Africa.
Afrobarometer Working Paper 111.
Krosnick, Jon A. 1999. Maximizing measurement quality: Principles of good questionnaire design. Eds.
John Paul Robinson, Phillip R. Shaver, and Lawrence S. Wrightsman. Measures of Political Attitudes. San
Diego: Academic Press.
Kull, Steven, Clay Ramsay, Stephen Weber, Evan Lewis. 2008. Pakistani and Indian Public Opinion on
Kashmir and Indo-Pakistani Relations. WorldPublicOpinion.Org, July 16, 2008. Available at:
http://www.worldpublicopinion.org/pipa/pdf/jul08/Kashmir_Jul08_rpt.pdf (accessed December
13, 2009).
Kull, Steven, Clay Ramsay, Stephen Weber, Evan Lewis , Ebrahim Mohseni. 2009. Public Opinion in the
Islamic World on Terrorism, al Qaeda, and US Policies. WorldPublicOpinin.org, February 25, 2009.
Available at: http://www.worldpublicopinion.org/pipa/pdf/feb09/STARTII_Feb09_rpt.pdf
(accessed December 13, 2009).
Lee, Eun Sul, and Ron N Forthofer. 2006. Analyzing Complex Survey Data. 2nd ed. Thousand Oaks, CA: Sage
Publications.
Levin, Carl. 2009. Opening Statement of Senator Carl Levin, Senate Armed Services Committee Hearing on
Afghanistan and Pakistan. February 26.
Marlowe, David, and Douglas P Crowne. 1964. The Approval Motive; Studies in Evaluative Dependence. New York:
Wiley.
Muller, Edward N. 1985. Income Inequality, Regime Repressiveness, and Political Violence. American
Sociological Review 50 (1): 47-61.
Musharraf, Pervez. 2006. In the Line of Fire: A Memoir. New York: Free Press.
Nasr, Vali R. 2000a. International Politics, Domestic Imperatives, and Identity Mobilization: Sectarianism in
Pakistan, 1979-1998. Comparative Politics 32 (2): 171-190.
Nasr ,Vali R. 2000b. The Rise of Sunni Militancy in Pakistan: The Changing Role of Islamism and the Ulama

34

in Society and Politics. Modern Asian Studies 34 (1): 139-180.

National Security Council of the United States. 2006 National Strategy for Combating Terrorism.
Persson, Torsten and Tabellini, Guido, 2009. "Democratic Capital: The Nexus of Political and Economic
Change." American Economic Journal: Macroeconomics 1(2): 88-126.
Pew Global Attitudes Project. 2002, 2004, 2005, 2006, 2007, 2008, 2009. Washington D.C.: Pew Research
Center. Available at: http://pewglobal.org/reports/ (accessed December 13, 2009).
Powell, G. Bingham. 1982. Contemporary Democracies: Participation, Stability, and Violence. Cambridge,
Massachusetts: Harvard University Press.
Rashid, Ahmed. 2000. Taliban: Militant Islam, Oil, and Fundamentalism in Central Asia. New Haven: Yale
University Press.
Shapiro, Jacob N. and C. Christine Fair. 2010. Understanding Support for Islamist Militancy in Pakistan.
International Security 34(3): 79-118.
Swami, Praveen. 2007. India, Pakistan and the Secret Jihad: The Covert War in Kashmi 19472004. London:
Routledge.
Terror Free Tomorrow, and New America Foundation. 2008. Results of a New Nationwide Public Opinion Survey
of Pakistan before the June 2008 Pakistani By-Elections. Washington D.C.: Terror Free Tomorrow and
New America Foundation. Available at: http://www.newamerica.net/files/TFT-Pakistan2008.pdf
(accessed December 13, 2009).
WorldPublicOpinion.org. 2009a. Pakistani Public Turns Against Taliban, But Still Negative on US.
WorldPublicOpinion.org: WorldPublicOpinion.org.
http://www.worldpublicopinion.org/pipa/pdf/jul09/WPO_Pakistan_Jul09_quaire.pdf (accessed
December 13, 2009).
WorldPublicOpinion.org. 2009b. Public Opinion in the Islamic World on Terrorism, al Qaeda, and US Policies.
Washington D.C.: WorldPublicOpinion.org.
http://www.worldpublicopinion.org/pipa/pdf/feb09/STARTII_Feb09_rpt.pdf (accessed
December 13, 2009).
Zaman, Muhammad Qasim. 1998. Sectarianism in Pakistan: The Radicalization of Shii and Sunni
Identities. Modern Asian Studies 32 (3): 689-716.

35

Appendix to Democratic Values and Support for Militancy: Evidence from a National
Survey of Pakistan
Online Appendix A: Sample Demographics
Gender
Male
Female

53.1%
46.9

Marital Status
Married
Not Married

77.8%
22.2

Urban/Rural
Urban
Rural

32.5%
67.5

Online Status
Online
Not Online

7.1%
92.3

Province
Punjab
Sindh
NWFP
Balochistan

55.6%
24.3
13.9
6.3

Own Cellphone
Yes
No

47.6%
52.4

Ability to Read
Yes
No

70.9%
29.1

Ability to Write
Yes
No

69.7%
30.3

Ability to Do Math
Yes
No

76.0%
24.0

Religious Sect
Sunni
Shiite
Age
18-24
25-29
30-39
40-49
50-59
60+

96.1%
3.9
22.9%
18.7
29.1
17.5
7.8
4.1

Education
Illiterate
Primary
Middle
Matriculant
Intermediate
Graduate
Professional

32.2%
13.1
14.9
19.3
12.3
6.4
1.9

Monthly Income
Less than 3000 PKR
3,000-10,000 PKR
10,001-15,000 PKR
15,001-25,000 PKR
More than 25,000 PKR

12.3%
53.9
22.9
8.8
2.2

Views on U.S. governments influence on world

Extremely positive
1.3%
Somewhat positive
6.0
Neither positive nor negative
8.7
Somewhat negative
20.9
Extremely negative
63.1
Views on U.S. governments influence on Pak.
Extremely positive
1.4%
Somewhat positive
6.0
Neither positive nor negative
9.8
Somewhat negative
18.2
Extremely negative
64.5
Believe that Sharia law is about physical punishment
Yes
55.0%
No
45.0

Note: N=6000 for all variables except monthly income (N=5779), views on U.S. governments influence on world (N=5859),
and views on U.S. governments influence on Pakistani politics (N=5874). Data weighted and adjusted for sampling design.

Online Appendix B: Question Wordings

Policies for Endorsement Experiment

The World Health Organizations recently announced a plan to introduced universal Polio vaccination across
Pakistan. How much do you support such a plan?
-A great deal
-A lot
-A moderate amount
-A little
-Not at all
The newly-elected national government has proposed reforming the Frontier Crimes Regulation and making
tribal areas equal to other provinces of the country. How much do you support such a plan?
-A great deal
-A lot
-A moderate amount
-A little
-Not at all
Governments of Pakistan and Afghanistan have explored using peace jirgas to resolve their disputes for
example the location of the boundary [Durand line/Sarhad]. How much do you support such a plan?
-A great deal
-A lot
-A moderate amount
-A little
-Not at all
In recent years the government of Pakistan has proposed curriculum reform for madaris to minimize
sectarian discord. How much do you support such a plan?
-A great deal
-A lot
-A moderate amount
-A little
-Not at all

Components of Kashmir Index

According to the non-governmental organization Human Rights Watch, Indian forces in Kashmir are
responsible for arbitrary detention, torture, and custodial killings of Kashmiri Muslims. How well does India
protect the rights of its Muslim citizens in Kashmir?
-Extremely well
-Somewhat well
-Neither well nor poorly
-Somewhat poorly
-Extremely poorly
How well does India protect the rights of its Muslim citizens?
-Extremely well
-Somewhat well
-Neither well nor poorly
-Somewhat poorly

ii

-Extremely poorly
According to the non-governmental organization Human Rights Watch, the Indian constitution provides for
freedom of religion for its Muslim citizens, and the National Government generally respected this right in the
last few years. How well does India protect the rights of its Muslim citizens in Kashmir?
-Extremely well
-Somewhat well
-Neither well nor poorly
-Somewhat poorly
-Extremely poorly
Thinking about the political preferences of Muslims in occupied Kashmir, please tell us which statement you
agree with the most:
-In occupied Kashmir the majority of Muslims want to be part of India.
-In occupied Kashmir the majority of Muslims want an independent state.
-In occupied Kashmir the majority of Muslims want to be part of Pakistan.

Religious Seriousness
If a child in your house were to study hafz-e-Quran or nazira, what kind of madrassa or school would you
like them to attend?
-Jamaat-e-Islami
-Ahl-e-hadith
-Deobandi
-Barlevi
-Shiite
-Ahle-sunnat
Do you attend dars-e-Quran?
-Yes
-No
How many times do you go to dars-e-Quran per week on average?

Jihadism
Some people say jihad is a personal struggle for righteousness. Others say jihad is protecting the Muslim
Ummah through war. What do you think?
-Jihad is solely a personal struggle for righteousness.
-Jihad is both a personal struggle for righteousness and protecting the Muslim Ummah through -war.
-Jihad is solely protecting the Muslim Ummah through war.
Some people say only a Muslim state/government can use military force to protect a Muslim country or
Ummah in the name of jihad. Others say individuals and non-state organizations can use military force in the
name of jihad. What do you think?
-Only states/governments should use military force in the name of jihad.
-Both states/governments and individuals should use military force in the name of jihad.
-Only individuals should use military force in the name of jihad.
-Neither can use military force in the name of jihad.

Democratic Values

iii

How important is it for you to live in a country that is governed by representatives elected by the people?
-Extremely important
-Very important
-Moderately important
-Slightly important
-Not important at all
How important is it for you to live in a country where the decisions of the courts are independent from
influence by political and military authorities?
-Extremely important
-Very important
-Moderately important
-Slightly important
-Not important at all
How important is it that individuals be able to express their political views, even though other people may
not agree with them?
-Extremely important
-Very important
-Moderately important
-Slightly important
-Not important at all
How important is it that individuals be able to meet with others to work on political issues?
-Extremely important
-Very important
-Moderately important
-Slightly important
-Not important at all
How important is it that individual property rights be secure? This means the state cannot take away their
things without proper court proceedings?
-Extremely important
-Very important
-Moderately important
-Slightly important
-Not important at all
The 1973 Constitution of Pakistan says civilians should control the military. This means the military cannot
take action without orders from civilian leaders. In your opinion, how much control should civilians have
over the military?
-Complete control
-A lot of control
-A moderate amount of control
-A little control
-No control at all
Tell us if you agree with this opinion: a Sharia government means a government that uses physical
punishments (stoning, cutting off hands, whipping) to make sure people obey the law.
-Agree
-Disagree

iv

Attitudes toward Pakistan Taliban, Afghan Taliban, and the United States Government
Were now going to ask you about a number of different groups. For each group, please answer to the best of
your ability. The first group is Pakistani militant groups fighting in Kashmir
What do you think is the groups objectives? Please tell us all that apply.
-Justice
-Democracy
-Protecting muslims
-Ridding the Muslim umma of people who have moved away from their religion
-Freeing Occupied Kashmir
Which government had a more legitimate claim to ruling Afghanistan: the current Karzai government or the
former Taliban government?
-The current Karzai government
-The former Taliban government
Which government performed better: the current Karzai government or the former Taliban government?
-The current Karzai government
-The former Taliban government
Here are two statements people make about the Afghan Taliban. Please tell us which you agree with more:
-The Afghan Taliban are trying to liberate Afghanistan from the United States and other foreign powers.
-The Afghan Taliban are rebelling against the government of Afghanistan.
Do you think the United States is occupying Afghanistan?
-Yes
-No

Negative Views of United States

Please tell us about the U.S. governments influence on the world, if it is: extremely positive, somewhat
positive, neither positive nor negative, somewhat negative, or extremely negative?
-Extremely positive
-Somewhat positive
-Neither positive nor negative
-Somewhat negative
-Extremely negative
Please tell us about the U.S. governments influence on Pakistans politics, if it is: extremely positive,
somewhat positive, neither positive nor negative, somewhat negative, or extremely negative?
-Extremely positive
-Somewhat positive
-Neither positive nor negative
-Somewhat negative
-Extremely negative

Demographics
Are you married?
-Yes
-No

Are you Sunni or Shiite?

-Sunni
-Shiite
-Non-Muslim [WRITTEN IN BY INTERVIEWER IF NON-MUSLIM]
What is your age in years?
What was the highest class you completed?
-Primary
-Middle
-Matriculant
-Intermediate (F.A/F.Sc)
-Graduate (B.A/B.Sc.)
-Professionals (M.S.C., M.A., Ph.D. or other professional degree)
-Illiterate
What is the approximate monthly income in your household?
-Less than 3000 rupees
-3000 to 10,000 rupees
-10,001 to 15,000 rupees
-15,001 to 25,000 rupees
-More than 25,000 rupees
Do you ever go online to access the Internet, do web site browsing, or to send and receive email?
-Yes
-No
Do you have a personal cell phone?
-Yes
-No
Read in any language with understanding?
-Yes
-No
-If yes, what language?
Can you write in any language, more than signing your name?
-Yes
-No
-If yes, what language?
Can you solve simple math (addition, subtraction) problems? Like 10 plus 7, or 30 divided by 5?
-Yes
-No
-Not sure
Sex:
-Male
-Female

vi

Online Appendix C: Randomization Checks

Control

Treatment

Gender
Male
Female
F: p=.99

53.1%
46.9

53.1%
46.9

Urban/Rural
Urban
Rural
F: p=.78

32.6%
67.4

32.3%
67.7

Province
Punjab
Sindh
NWFP
Balochistan
F: p=.72

55.8%
24.4
13.5
6.3

55.3%
24.1
14.3
6.2

Religious Sect
Sunni
Shiite
F: p=.51

96.3%
3.7

96.0%
4.0

Age
18-24
25-29
30-39
40-49
50-59
60+
F: p=.63

23.2%
19.3
28.2
17.5
7.9
3.9

22.5%
18.2
29.9
17.4
7.6
4.3

Education
Illiterate
Primary
Middle
Matriculant
Intermediate
Graduate
Professional
F: p=.21

32.4%
13.7
13.8
19.6
12.8
6.1
1.7

31.9%
12.6
15.9
19.0
11.9
6.7
2.1

Monthly Income
Less than 3000 PKR
3,000-10,000 PKR
10,001-15,000 PKR
15,001-25,000 PKR
More than 25,000 PKR
F: p=.31

9.7%
55.8
23.9
8.8
1.8

9.6%
53.6
24.7
9.8
2.2

Note: N=6000 for all variables except monthly income

(N=5779). Data weighted and adjusted for sampling design.

vii